Re: [ietf-smtp] Make username optional in email addresses

[Russ Allbery <[e-mail]>]

Dave Crocker <[e-mail]> writes:
> On 2/17/2023 1:45 PM, Russ Allbery wrote:

>> I think a comparable change would be MIME, which introduced structured
>> email and non-ASCII character sets, and that change took decades to be
>> reliably usable

> MIME is a likely guess for comparison, but it's actually fundamentally
> different.  It did not touch the infrastructure. MIME is a pure
> value-add function that rides on top of the existing email handling and
> content standards.  In terms of pure computer science, it is arguably
> ugly but in terms of demonstrating how to upgrade a global system, it's
> been astonishingly effective engineering.

I completely agree, and indeed that was the point of the rest of the
paragraph that you didn't quote.  :)  But I probably didn't make that
clear enough.

> And while I don't have measurements, I think it became useful far more
> quickly than decades.

Also agreed, but the term I used was "reliably useful" not just "useful."
What I was trying to get at, without enough elaboration, is that it takes
a really long time to upgrade enough software for a new feature to become
something people can *rely* on working, particularly with a stranger with
no prior negotiation.

That's relevant in the choice in, say, whether you feel comfortable giving
out a domain name as your email address and being confident any email
sender you care about will use it.  And providing your email address to
people who don't already have it feels like the compelling use case of
this feature.

(That said, email is a lot more centralized these days, and if Gmail and
Outlook both implement something, you get a lot of coverage very, very
quickly.  I'm not sure they would be interested in this feature, though.)

-- 
Russ Allbery ([e-mail])             <https://www.eyrie.org/~eagle/>